Mid Sweden University

The purpose of the present study was threefold: (a) to replicate the finding by MacLeod and Rutherford (1992) that low trait-anxious individuals, in contrast to high trait-anxious individuals, show a selective attention away from subliminally presented threat words at elevated levels of stress; (b) to test the hypothesis that this effect is due to individuals with repressive coping-style rather than true low trait-anxious individuals; and (c) to study the stability of Stroop interference over time. Both social threat and physical threat words were used. Although some support was found for the first hypothesis, there was no evidence that this effect was due to individuals with repressive coping-style. Finally, Stroop interference showed very little test–retest stability from the first to the second testing session, indicating that it is heavily influenced by temporary cognitive-emotional states, and that it should not be treated as a trait variable. Unexpectedly, high defensiveness predicted a decrease in worried mood from session 1 to session 2.

The role of defensiveness and repressive coping style for the performance on a combined emotional Stroop and tachistoscopic identification task with masked and unmasked words was studied in a community sample. Defensiveness was associated with a decrease in Stroop interference for masked threat words, but not for unmasked threat words. The most robust results, however, were found with regard to overall test performance (independent of emotional valence). On the emotional Stroop task, high-defensive men (but not women) were faster to colour-name words in general, irrespective of emotional valence. On the tachistoscopic identification task, high-defensive women identified fewer words in general than low-defensive participants. The results are discussed in terms of defensiveness being associated with (a) avoidance of emotional information at an automatic, pre-attentive level, and (b) a general avoidance of potentially emotional information that takes different form in men and women depending on possible differences in what is seen as socially desirable for the two genders.

The associations between perfectionism, anger, somatic health, and positive affect were examined in 184 Swedish adults from a randomly selected population sample. Somewhat unexpectedly, trait anger was found to be associated with self-oriented perfectionism rather than with socially prescribed perfectionism. Both socially prescribed perfectionism and self-oriented perfectionism showed weak positive correlations with self-reported somatic complaints, particularly symptoms of tension and fatigue, and more clearly in women than in men, whereas other-oriented perfectionism appeared as a predictor of whether the participants were undergoing medical treatment or not. Finally, the results did not support the notion of self-oriented perfectionism representing a positive, adaptive dimension of perfectionism; on the contrary, this dimension was found to be negatively associated with positive affect. 

The present study replicates and extends Harvey, Clark, Ehlers and Rapee's single-session experiment of video feedback with cognitive preparation in individuals with high social anxiety. Forty participants high on social anxiety where asked to present a speech before a video camera and then received video feedback of their speech. Half of the participants received cognitive preparation prior to the video feedback, whereas the other half only received video feedback. The cognitiv preparation procedure asked the participants to (1) predict in detail what the yould see in the video, (2) form an image of themselves giving the speech, and (3) watch the video as though they were watching a stranger. All the participants were asked to present a second speech 30 min after the first speech to index generalisation effects. Cognitive preparation enchanced the therapeutic effect of video feedback after the first speech, and also showed a robust gneralisation effect on the evaluation of the second speech. Beofred the second speech, the participants underwent an emotional Stroop task with masked and unmasked words; the results showed that Stroop interference for both masked and unmasked social threat words correlated consistently with measures of social anxiety, but not with the preceding or ensuing speech evaluations.